Agent Memory MCP
未认领MCP server that gives AI agents persistent memory with semantic search
安装
go install github.com/ipiton/agent-memory-mcp/cmd/agent-memory-mcp@latest更多Search & Web服务器
浏览完整目录安全概况
已认领和已认证的服务器每周接受一次静态扫描，显示代码能触及的范围（外部服务、环境变量、shell 命令、代理配置目录）以及带有已知公告的依赖。认领此列表即可获得。 安全概况的工作原理
40 个工具中显示 40 个
文档中列出的工具 (40)
内容来自项目文档。服务器握手不会验证每个工具的说明或行为。
accept_session_changes
Persist the raw summary and auto-apply only low-risk consolidation actions
analyze_session
Compatibility alias for closesession with the same planning and reporting behavior
close_session
Analyze a finished session into raw summary metadata, candidate knowledge items, and review-safe consolidation actions
conflicts_report
Report duplicate candidates, conflicting statuses, and multiple canonical entries
delete_memory
Delete a memory by ID
demote_sediment
Demote a memory one sediment layer down
end_task
Consolidate memory for an archived task slug: outdate working/procedural entries, route high-importance ones to the review queue
index_documents
Re-index documents for RAG search
list_canonical_knowledge
List canonical knowledge entries projected from confirmed memories
list_memories
List all memories with optional type/context filtering
mark_outdated
Mark a memory as outdated or superseded so trust-aware recall downranks it
memory_stats
Get memory statistics (counts by type)
merge_duplicates
Merge duplicate memories into a primary entry and archive the rest
project_bank_view
Show a structured project bank view for canonical knowledge, decisions, runbooks, incidents, caveats, migrations, the review queue, or sediment promotion candidates
promote_sediment
Promote a memory to a higher sediment layer (surface → episodic → semantic → character). See docs/SEDIMENTATION.md
promote_to_canonical
Promote a memory to canonical knowledge and boost its trust ranking
recall_canonical_knowledge
Recall canonical knowledge only, excluding raw memories from results
recall_memory
Recall memories by semantic/text query with optional filters and trust-aware ranking
recall_multihop
Multi-hop graph-walk recall over the (subj, rel, obj) triple corpus — returns memories ranked by aggregated path score with the chain of triples that reached each result. Use for cross-memory reasoning queries that single-hop search cannot trace. Requires MCPTRIPLEEXTRACTOR populated; backfill via i
recall_similar_incidents
Recall similar incidents from memory and indexed postmortems
repo_list
List files and folders under allowlisted paths
repo_read
Read a file from allowlisted paths
repo_search
Text search across allowlisted paths
resolve_review_item
Resolve a pending review queue item so it disappears from the active inbox while keeping an audit trail
review_session_changes
Render the explainable review report for a finished session without forcing writes
search_runbooks
Search runbook memories plus indexed runbook docs
sediment_cycle
Run the sediment transition cycle — auto-applies trivial promotions, routes non-trivial ones to the review queue
semantic_search
Hybrid search across indexed documents with optional sourcetype, trust metadata, and debug explain mode
steward_policy
Get or update the stewardship policy that controls detection thresholds, auto-apply rules, and scheduling
steward_report
Retrieve the latest stewardship report or a specific one by run ID
steward_run
Run a knowledge stewardship cycle: scan for duplicates, conflicts, stale entries, and canonical promotion candidates
store_dead_end
Record an attempted approach that failed (plus the why and the alternative used) so retrieval can surface it as a pitfall warning on related queries. Use this for standalone failures with no decision context. Use storedecision -avoided-dead-end-id <id> when the dead end is part of a larger architect
store_decision
Store an engineering decision with rationale, status, and consequences
store_incident
Store an incident with impact, root cause, resolution, service, and severity
store_memory
Store a memory with content, type, tags, and importance
store_postmortem
Store a postmortem with root cause and action items
store_runbook
Store a runbook with procedure, trigger, verification, and rollback notes
summarize_project_context
Summarize recent decisions, runbooks, incidents, and related docs
sweep_archive
Pull-mode scan over MCPTASKARCHIVEROOTS that runs endtask on every archived slug
update_memory
Update an existing memory by ID